Tree species recognition in species rich area using UAV-borne hyperspectral imagery and stereo-photogrammetric point cloud

S. Tuominen, R. Näsi, E. Honkavaara, A. Balazs, T. Hakala, N. Viljanen, I. Pölönen, H. Saari, J. Reinikainen

    Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

    1 Citation (Scopus)

    Abstract

    Recognition of tree species and geospatial information of tree species composition is essential for forest management. In this study we test tree species recognition using hyperspectral imagery from VNIR and SWIR camera sensors in combination with 3D photogrammetric canopy surface model based on RGB camera stereo-imagery. An arboretum forest with a high number of tree species was used as a test area. The imagery was acquired from the test area using UAV-borne cameras. Hyperspectral imagery was calibrated for providing a radiometrically corrected reflectance mosaic, which was tested along with the original uncalibrated imagery. Alternative estimators were tested for predicting tree species and genus, as well as for selecting an optimal set of remote sensing features for this task. All tested estimators gave similar trend in the results: the calibrated reflectance values performed better in predicting tree species and genus compared to uncorrected hyperspectral pixel values. Furthermore, the combination of VNIR, SWIR and 3D features performed better than any of the data sets individually, with calibrated reflectances and original pixel values alike. The highest proportion of correctly classified trees was achieved using calibrated reflectance features from VNIR and SWIR imagery together with 3D point cloud features: 0.823 for tree species and 0.869 for tree genus.

    Original languageEnglish
    Title of host publicationFrontiers in Spectral imaging and 3D Technologies for Geospatial Solutions
    EditorsE. Honkavaara, B. Hu, K. Karantzalos, X. Liang, R. Müller, E. Nocerino, I. Pölönen, P. Rönnholm
    PublisherInternational Society for Photogrammetry and Remote Sensing ISPRS
    Pages185-194
    Number of pages10
    DOIs
    Publication statusPublished - 2017
    MoE publication typeA4 Article in a conference publication
    EventFrontiers in Spectral imaging and 3D Technologies for Geospatial Solutions, ISPRS SPEC3D - Jyväskylä, Finland
    Duration: 25 Oct 201727 Oct 2017
    http://www.mit.jyu.fi/scoma/spec3d/
    http://www.mit.jyu.fi/scoma/spec3d/
    http://www.mit.jyu.fi/scoma/spec3d/

    Publication series

    SeriesInternational Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
    VolumeXLII-3/W3
    ISSN1682-1750

    Workshop

    WorkshopFrontiers in Spectral imaging and 3D Technologies for Geospatial Solutions, ISPRS SPEC3D
    Abbreviated titleISPRS SPEC3D
    CountryFinland
    CityJyväskylä
    Period25/10/1727/10/17
    Internet address

    Fingerprint

    imagery
    reflectance
    pixel
    forest management
    canopy
    sensor
    remote sensing
    test

    Keywords

    • hyperspectral imaging
    • UAVs
    • stereo-photogrammetry
    • photogrammetric point cloud
    • tree species recognition

    Cite this

    Tuominen, S., Näsi, R., Honkavaara, E., Balazs, A., Hakala, T., Viljanen, N., ... Reinikainen, J. (2017). Tree species recognition in species rich area using UAV-borne hyperspectral imagery and stereo-photogrammetric point cloud. In E. Honkavaara, B. Hu, K. Karantzalos, X. Liang, R. Müller, E. Nocerino, I. Pölönen, ... P. Rönnholm (Eds.), Frontiers in Spectral imaging and 3D Technologies for Geospatial Solutions (pp. 185-194). International Society for Photogrammetry and Remote Sensing ISPRS. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol.. XLII-3/W3 https://doi.org/10.5194/isprs-archives-XLII-3-W3-185-2017
    Tuominen, S. ; Näsi, R. ; Honkavaara, E. ; Balazs, A. ; Hakala, T. ; Viljanen, N. ; Pölönen, I. ; Saari, H. ; Reinikainen, J. / Tree species recognition in species rich area using UAV-borne hyperspectral imagery and stereo-photogrammetric point cloud. Frontiers in Spectral imaging and 3D Technologies for Geospatial Solutions. editor / E. Honkavaara ; B. Hu ; K. Karantzalos ; X. Liang ; R. Müller ; E. Nocerino ; I. Pölönen ; P. Rönnholm. International Society for Photogrammetry and Remote Sensing ISPRS, 2017. pp. 185-194 (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. XLII-3/W3).
    @inproceedings{6f6f5542e3f74a5e8459e882bdea180c,
    title = "Tree species recognition in species rich area using UAV-borne hyperspectral imagery and stereo-photogrammetric point cloud",
    abstract = "Recognition of tree species and geospatial information of tree species composition is essential for forest management. In this study we test tree species recognition using hyperspectral imagery from VNIR and SWIR camera sensors in combination with 3D photogrammetric canopy surface model based on RGB camera stereo-imagery. An arboretum forest with a high number of tree species was used as a test area. The imagery was acquired from the test area using UAV-borne cameras. Hyperspectral imagery was calibrated for providing a radiometrically corrected reflectance mosaic, which was tested along with the original uncalibrated imagery. Alternative estimators were tested for predicting tree species and genus, as well as for selecting an optimal set of remote sensing features for this task. All tested estimators gave similar trend in the results: the calibrated reflectance values performed better in predicting tree species and genus compared to uncorrected hyperspectral pixel values. Furthermore, the combination of VNIR, SWIR and 3D features performed better than any of the data sets individually, with calibrated reflectances and original pixel values alike. The highest proportion of correctly classified trees was achieved using calibrated reflectance features from VNIR and SWIR imagery together with 3D point cloud features: 0.823 for tree species and 0.869 for tree genus.",
    keywords = "hyperspectral imaging, UAVs, stereo-photogrammetry, photogrammetric point cloud, tree species recognition",
    author = "S. Tuominen and R. N{\"a}si and E. Honkavaara and A. Balazs and T. Hakala and N. Viljanen and I. P{\"o}l{\"o}nen and H. Saari and J. Reinikainen",
    year = "2017",
    doi = "10.5194/isprs-archives-XLII-3-W3-185-2017",
    language = "English",
    series = "International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences",
    publisher = "International Society for Photogrammetry and Remote Sensing ISPRS",
    pages = "185--194",
    editor = "E. Honkavaara and B. Hu and K. Karantzalos and X. Liang and R. M{\"u}ller and E. Nocerino and I. P{\"o}l{\"o}nen and P. R{\"o}nnholm",
    booktitle = "Frontiers in Spectral imaging and 3D Technologies for Geospatial Solutions",

    }

    Tuominen, S, Näsi, R, Honkavaara, E, Balazs, A, Hakala, T, Viljanen, N, Pölönen, I, Saari, H & Reinikainen, J 2017, Tree species recognition in species rich area using UAV-borne hyperspectral imagery and stereo-photogrammetric point cloud. in E Honkavaara, B Hu, K Karantzalos, X Liang, R Müller, E Nocerino, I Pölönen & P Rönnholm (eds), Frontiers in Spectral imaging and 3D Technologies for Geospatial Solutions. International Society for Photogrammetry and Remote Sensing ISPRS, International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. XLII-3/W3, pp. 185-194, Frontiers in Spectral imaging and 3D Technologies for Geospatial Solutions, ISPRS SPEC3D, Jyväskylä, Finland, 25/10/17. https://doi.org/10.5194/isprs-archives-XLII-3-W3-185-2017

    Tree species recognition in species rich area using UAV-borne hyperspectral imagery and stereo-photogrammetric point cloud. / Tuominen, S.; Näsi, R.; Honkavaara, E.; Balazs, A.; Hakala, T.; Viljanen, N.; Pölönen, I.; Saari, H.; Reinikainen, J.

    Frontiers in Spectral imaging and 3D Technologies for Geospatial Solutions. ed. / E. Honkavaara; B. Hu; K. Karantzalos; X. Liang; R. Müller; E. Nocerino; I. Pölönen; P. Rönnholm. International Society for Photogrammetry and Remote Sensing ISPRS, 2017. p. 185-194 (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. XLII-3/W3).

    Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

    TY - GEN

    T1 - Tree species recognition in species rich area using UAV-borne hyperspectral imagery and stereo-photogrammetric point cloud

    AU - Tuominen, S.

    AU - Näsi, R.

    AU - Honkavaara, E.

    AU - Balazs, A.

    AU - Hakala, T.

    AU - Viljanen, N.

    AU - Pölönen, I.

    AU - Saari, H.

    AU - Reinikainen, J.

    PY - 2017

    Y1 - 2017

    N2 - Recognition of tree species and geospatial information of tree species composition is essential for forest management. In this study we test tree species recognition using hyperspectral imagery from VNIR and SWIR camera sensors in combination with 3D photogrammetric canopy surface model based on RGB camera stereo-imagery. An arboretum forest with a high number of tree species was used as a test area. The imagery was acquired from the test area using UAV-borne cameras. Hyperspectral imagery was calibrated for providing a radiometrically corrected reflectance mosaic, which was tested along with the original uncalibrated imagery. Alternative estimators were tested for predicting tree species and genus, as well as for selecting an optimal set of remote sensing features for this task. All tested estimators gave similar trend in the results: the calibrated reflectance values performed better in predicting tree species and genus compared to uncorrected hyperspectral pixel values. Furthermore, the combination of VNIR, SWIR and 3D features performed better than any of the data sets individually, with calibrated reflectances and original pixel values alike. The highest proportion of correctly classified trees was achieved using calibrated reflectance features from VNIR and SWIR imagery together with 3D point cloud features: 0.823 for tree species and 0.869 for tree genus.

    AB - Recognition of tree species and geospatial information of tree species composition is essential for forest management. In this study we test tree species recognition using hyperspectral imagery from VNIR and SWIR camera sensors in combination with 3D photogrammetric canopy surface model based on RGB camera stereo-imagery. An arboretum forest with a high number of tree species was used as a test area. The imagery was acquired from the test area using UAV-borne cameras. Hyperspectral imagery was calibrated for providing a radiometrically corrected reflectance mosaic, which was tested along with the original uncalibrated imagery. Alternative estimators were tested for predicting tree species and genus, as well as for selecting an optimal set of remote sensing features for this task. All tested estimators gave similar trend in the results: the calibrated reflectance values performed better in predicting tree species and genus compared to uncorrected hyperspectral pixel values. Furthermore, the combination of VNIR, SWIR and 3D features performed better than any of the data sets individually, with calibrated reflectances and original pixel values alike. The highest proportion of correctly classified trees was achieved using calibrated reflectance features from VNIR and SWIR imagery together with 3D point cloud features: 0.823 for tree species and 0.869 for tree genus.

    KW - hyperspectral imaging

    KW - UAVs

    KW - stereo-photogrammetry

    KW - photogrammetric point cloud

    KW - tree species recognition

    UR - http://www.scopus.com/inward/record.url?scp=85033693009&partnerID=8YFLogxK

    U2 - 10.5194/isprs-archives-XLII-3-W3-185-2017

    DO - 10.5194/isprs-archives-XLII-3-W3-185-2017

    M3 - Conference article in proceedings

    T3 - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences

    SP - 185

    EP - 194

    BT - Frontiers in Spectral imaging and 3D Technologies for Geospatial Solutions

    A2 - Honkavaara, E.

    A2 - Hu, B.

    A2 - Karantzalos, K.

    A2 - Liang, X.

    A2 - Müller, R.

    A2 - Nocerino, E.

    A2 - Pölönen, I.

    A2 - Rönnholm, P.

    PB - International Society for Photogrammetry and Remote Sensing ISPRS

    ER -

    Tuominen S, Näsi R, Honkavaara E, Balazs A, Hakala T, Viljanen N et al. Tree species recognition in species rich area using UAV-borne hyperspectral imagery and stereo-photogrammetric point cloud. In Honkavaara E, Hu B, Karantzalos K, Liang X, Müller R, Nocerino E, Pölönen I, Rönnholm P, editors, Frontiers in Spectral imaging and 3D Technologies for Geospatial Solutions. International Society for Photogrammetry and Remote Sensing ISPRS. 2017. p. 185-194. (International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol. XLII-3/W3). https://doi.org/10.5194/isprs-archives-XLII-3-W3-185-2017